Credit markets, financial crises, and the macroeconomy.

摘要

This study consists of three chapters, each of which is an individual paper. The first chapter investigates how the dynamic process of reallocation of credit across firms behaves before and after financial crises. Applying the methodology proposed by Davis and Haltiwanger (1992) for measuring job reallocation, we construct measures of credit reallocation across Korean firms in the 1981-2012 period. The credit boom preceding the 1997 financial crisis featured a modest intensity of credit reallocation. By contrast, after the crisis and the associated reforms, credit reallocation significantly intensified and started to comove with the business cycle, while credit growth slowed down (deleveraging). The higher dynamism of the credit sector in reallocating liquidity cannot be explained by "flight to quality" episodes but reflects a structural change in the credit reallocation process that has persisted since the end of the crisis. The intensification of credit reallocation appears to have been associated with enhanced allocative efficiency.;The second chapter explores the evolution of credit reallocation across Korean non-financial firms for the period 1981-2012. I employ a dynamic latent factor model that decomposes regional credit reallocation rates into national, region-specific and idiosyncratic components. I find that the common factor explaining common movement across 16 regional credit flows increased after the 1997 financial crisis. The common factor comoves with national excess reallocation. It is positively and strongly correlated with national excess reallocation, while it is negatively correlated with national net credit growth. It exhibits mild counter-cyclicality. I examine what extent the volatility of credit reallocation was driven by national, region-specific and idiosyncratic components. This study uncovers evidence that the national factor accounts for a sizable fraction of regional reallocation rates of total credit and loans, while it plays only a minor role in explaining the fluctuation in regional reallocation rates of bonds.;The last chapter explores the relationship between religion and bank performance. The study uses data on credit unions in Korea for the period 2000 to 2007 to investigate the effects of religion on bank performance. The empirical results show that credit unions based on religious institutions not only suffer less from troubled loans but they also enjoy higher profits relative to ordinary ones. I find that the religious credit unions' unique features, such as non random potential clientele, rich soft information and reputational incentive to repay, are likely to be what enables them to outperform.